Coherent load monitoring of physical and virtual networks with synchronous status acquisition

Abstract

Embodiments relate to creating a coherent load or congestion map that displays the simultaneous activity of all queues of physical and virtual switches and adapters in a network without forcing clock synchronization. An aspect includes sampling, by a central processing device, a status of all queues in a plurality of elements in a network. The sampled data flows are received from the plurality of elements in the network and an image is created of the network. The image includes the status of all sampled queues in the plurality of elements at a point in time. Accordingly, a load map is created without synchronizing clocks of the plurality of elements. The load map is assembled using segments of the image of the network.

What is claimed is: 
     
       1. A computer-implemented method, comprising:
 sampling, by a central processing device, a status of all queues in a network, the queues located on a plurality of elements in the network; 
 receiving the sampled status of all queues from the plurality of elements; 
 creating an image of the network, the image including the status of all sampled queues in the plurality of elements at a point in time; and 
 creating a load map without synchronizing clocks of the plurality of elements, the load map being assembled using segments of the image of the network, wherein the central controller and elements are part of a software-defined networking environment. 
 
     
     
       2. The computer-implemented method of  claim 1 , wherein the creating of a load map further comprises handing off the image to another server to create the load map offline. 
     
     
       3. The computer-implemented method of  claim 1 , wherein the load map is displayed as a heat map. 
     
     
       4. The computer-implemented method of  claim 1 , wherein the sampling of the network further comprises dynamically varying the sampling frequency time scale based on detected network congestion. 
     
     
       5. The computer-implemented method of  claim 1 , wherein the sampling of the network further is triggered by an event comprising at least one of an application, a network event, or an OpenFlow console. 
     
     
       6. The computer-implemented method of  claim 1 , further comprising:
 creating subsequent load maps; 
 identifying hotspots in an evolution of the network based on the subsequent load maps; and 
 sampling areas of the network that are prone to congestion more frequently. 
 
     
     
       7. The computer-implemented method of  claim 1 , further comprising:
 creating subsequent load maps; 
 utilizing historical data to identify new anomalies in the network based on the subsequent load maps; and 
 sampling areas of the network showing new anomalies more frequently. 
 
     
     
       8. The computer-implemented method of  claim 1 , wherein creating the load map further comprises:
 handing off the image to a server to create the load map offline; and 
 receiving the load map from the server, the load map created without synchronizing clocks of the plurality of elements from the server, the load map being assembled using segments of the image of the network. 
 
     
     
       9. A computer system, comprising:
 a memory having computer readable instructions; and a processing device for executing the computer readable instructions, the instructions including: 
 sampling, by a central processing device, a status of all queues in a network, the queues located on a plurality of elements in the network; 
 receiving the sampled data flows from the plurality of elements; 
 creating an image of the network, the image including the status of all sampled queues in the plurality of elements at a point in time; and 
 creating a load map without synchronizing clocks of the plurality of elements, the load map being assembled using segments of the image of the network, wherein the central controller and elements are part of a software-defined networking environment. 
 
     
     
       10. The computer system of  claim 9 , wherein the creating of a load map further comprises handing off the image to another server to create the load map offline. 
     
     
       11. The computer system of  claim 9 , wherein the load map is displayed as an infrared heat map. 
     
     
       12. The computer system of  claim 9 , wherein the sampling of the network further comprises dynamically varying the sampling frequency time scale based on detected network congestion. 
     
     
       13. The computer system of  claim 9 , wherein the sampling of the network further is triggered by an event comprising at least one of an application, a network event, or an OpenFlow console. 
     
     
       14. The computer system of  claim 9 , further comprising:
 creating subsequent load maps; 
 identifying hotspots in an evolution of the network based on the subsequent load maps; and 
 sampling areas of the network that are prone to congestion more frequently. 
 
     
     
       15. The computer system of  claim 9 , further comprising:
 creating subsequent load maps; 
 utilizing historical data to identify new anomalies in the network based on the subsequent load maps; and 
 sampling areas of the network showing new anomalies more frequently. 
 
     
     
       16. A computer program product, comprising:
 a computer readable storage medium wherein the medium is not a signal and wherein the medium having program code embodied therewith, the program code executable by a processor for: 
 sampling, by a central processing device, a status of all the queues in a network, the queues located on a plurality of elements in the network; 
 receiving the sampled status of all queues from the plurality of elements; 
 creating an image of the network, the image including the status of all sampled queues in the plurality of elements at a point in time; and 
 creating a load map without synchronizing clocks of the plurality of elements, the load map being assembled using segments of the image of the network, wherein the central controller and elements are part of a software-defined networking environment. 
 
     
     
       17. The computer program product of  claim 16 , wherein the creating of a load map further comprises handing off the image to another server to create the load map offline. 
     
     
       18. The computer program product of  claim 16 , wherein the load map is displayed as an infrared heat map. 
     
     
       19. The computer program product of  claim 16 , wherein the sampling of the network further comprises dynamically varying the sampling frequency time scale based on detected network congestion. 
     
     
       20. The computer program product of  claim 16 , wherein the sampling of the network further is triggered by an event comprising at least one of an application, a network event, or an OpenFlow console. 
     
     
       21. The computer program product of  claim 16 , further comprising:
 creating subsequent load maps; 
 identifying hotspots in an evolution of the network based on the subsequent load maps; and 
 sampling areas of the network that are prone to congestion more frequently. 
 
     
     
       22. The computer program product of  claim 16 , further comprising:
 creating subsequent load maps; 
 utilizing historical data to identify new anomalies in the network based on the subsequent load maps; and 
 sampling areas of the network showing new anomalies more frequently.